Neuropsychological and Neuroimaging Findings of Semantic Dementia

BACKGROUND: Semantic dementia (SD) is a temporal variant of frontotemporal lobar degeneration (FTLD), which is characterized by naming difficulty, decreased comprehension of words, prosopagnosia and object visual agnosia. We report clinical features, neuropsychological and neuroimaging findings of 4 patients with SD. METHODS: Of 55 patients diagnosed as having FTLD between Jan 1995 and May 2001 at Samsung Medical Center, four patients fulfilled the diagnostic criteria of SD proposed by consensus on FTLD diagnostic criteria. We investigated their clinical features such as presenting symptoms and abnormal behaviors, neuropsychological and neuroimaging findings. Neuropsychological tests included the Seoul Neuropsychological Screening Battery, the Korean-version of Western Aphasia Battery and Hanja reading and writing. All patients underwent brain MRI and FDG-PET. RESULTS: All of the patients showed naming difficulty as a presenting symptom. Language assessments showed severe naming and compre-hension difficulties with preserved fluency and repetition, which were compatible with transcortical sensory aphasia. Whereas Hangul reading aloud and writing were intact, three patients were impaired at Hanja reading and writing. Other neuropsychological tests were remarkable for prosopagnosia, object visual agnosia and memory loss. Brain MRI showed asymmetric temporal atrophies, mainly left antero-inferior temporal cortices. FDG-PET also showed hypome-tabolism in bilateral anterior temporal lobes, more severe on the left. CONCLUSIONS: Our SD patients had characteristic neuropsychological and neuroimaging findings, which can be differentiated from other neurodegenerative diseases. We also found Hanja alexia and agraphia in SD patients, which has not been reported yet.

The Predictive Value of FDG-PET for Lateralizing Wada Memory Dominance and Epileptic Focus in Patients with Temporal Lobe Epilepsy

Backgroud : To examine the predictability of regional cerebral glucose metabolism in determining Wada memory dominance and lateralizing epileptic focus. METHODS: 1 8 F-fluorodeoxyglucose positron emission tomography (FDG-PET) and Wada test were performed in 18 patients with temporal lobe epilepsy (TLE). Regions of interest were determined in mesial, polar, anterior-lateral, mid-lateral, and posterior-lateral regions of the temporal lobe. The asymmetry indices of FDG-PET (PET-AI) were calculated in each ROI of temporal lobe, and those of Wada memory test (Wada-AI) were obtained as well. RESULTS: Pearson correlation coefficient showed Wada-AI was significantly correlated with PET-AI in mesial (r=0.67, p=0.001), polar (r=0.55, p=0.010), anterior-lateral (0.55, p=0.009) and mid-lateral (r=0.51, p=0.016) temporal regions. However, after simple linear regression analysis, PET-AI of mesial temporal region alone was significantly correlated with Wada-AI (p=0.008). In localizing epileptic focus, Wada-AI could correctly lateralize the seizure focus in 90% of the left TLE and 75% of the right TLE patients. No false lateralization by Wada-AI was observed except two patients showing prolonged confusion after amobarbital injection who were not included in this study. The PET-AI of the mesial temporal region showed the highest sensitivity of seizure lateralization (100% of left TLE and 87.5% of right RLE). CONCLUSIONS: Although FDG-PET hypometabolism is observed both at mesial and lateral regions of the temporal lobe in mesial TLE, mesial temporal region appeared to be a dominant and leading area for lateralizing Wada memory dominance and epileptic focus.

The Extent of Hippocampal Resection and Postsurgical Memory Change in Temporal Lobe Epilepsy / 대한간질학회지

BACKGROUND: This study was performed to investigate the relationship between the extent of hippocampal resection and the postsurgical memory outcome in temporal lobe epilepsy (TLE). METHODS: In 36 patients with TLE, the longitudinal distances of pre-surgical hippocampus and post-surgical hippocampal remnant were measured on 3 mm thick coronal MRI images perpendicular to the long axis of hippocampus. Memory tests were performed before and 1 year after the surgery. The relationships of the extent of hippocampal resection (EHR) and the asymmetry index of Wada retention memory scores (WAI) with postsurgical memory outcomes were tested. RESULTS: The ratios of post-surgical/pre-surgical scores in immediate and delayed verbal memory and immediate, delayed and recognition visual memory were not significantly correlated with the EHR. Only verbal recognition memory was positively correlated with the EHR. In TLE of the dominant hemisphere, the ratios of post-surgical/pre-surgical scores of verbal and visual memories were not significantly correlated with the EHR, but the WAIs were significantly correlated with the delayed visual memory changes (p<0.05). In TLE of the non-dominant hemisphere, the ratios of post-surgical/pre-surgical scores of delayed verbal and immediate visual memory (in percentiles) were positively correlated with the EHR (p<0.05). However, a linear regression analysis showed that none of postsurgical memory subtypes were significantly correlated with the EHR. CONCLUSIONS: This study suggests that the extent of hippocampal resection itself does not have a significant relationship with the outcome of postsurgical memory in patients with mesial TLE.

Clinical Factors Influencing Quality of Life in Patients with Epilepsy

BACKGROUND: Epilepsy has been known to adversely affect a patient's quality of life (QOL). We investigated the status of QOL and tested the influences of clinical factors upon QOL in epileptic patients. METHODS: We evaluated 125 adult patients (mean age, 30.4 years, male:female=1:0.86) with epilepsy. The demographic, social (marriage, employment, religion, education duration), and clinical data (seizure onset age, seizure pattern during the last 1 year, treatment duration, number of antiepileptic drug) were obtained. Quality of life in epilepsy-31 (QOLIE-31) and Beck Depression Inventory-Korean version (KBDI) were used. The relationships of demographic, social, and clinical factors with QOL in epileptic patients were evaluated. RESULTS: The mean subscores of QOLIE-31 items were 52.6 (medication effects), 50.9 (cognitive functioning), 50.9 (social functioning), 50.5 (energy/fatigue), 48.4 (seizure worry), 46.8 (overall QOL), and 45.7 (emotional well-being), and the overall score was 49.6. Sex, marriage, religion, and duration of treatment did not affect QOL. The age of patients had a weak linear positive correlation with overall QOL and emotional well-being. Employed patients had a significantly high score in overall QOL, emotional well-being, cognitive functioning, social functioning, and overall score. The patients given polytherapy had lower scores of QOLIE-31 items except seizure worry and energy/fatigue. Seizure free patients had significantly higher scores in seizure worry, overall QOL, social functioning, and overall score. Depression negatively influenced all subscales and overall score. CONCLUSIONS: Age, employment state, number of antiepileptic drugs, seizure pattern, and depression were significant clinical factors affecting QOL in patients with epilepsy.

Right Anterior Thalamic Infarction Presenting Only with Cognitive Dysfunction

A 60 year-old right-handed man developed a sudden inability to find his way home from work. Neurologic examina-tions revealed no focal neurological deficit. Neuropsychological tests showed a memory deficit especially in the visu-ospatial domain along with visuospatial and frontal-executive dysfunction. A brain MRI revealed a small but discrete lesion in the right thalamus. The majority of the lesion was confined to the dorsomedial nucleus. These results suggest that a small unilateral thalamic infarction affecting the anterior area may be sufficient to produce "strategic-infarct dementia" without focal neurological deficit.

Right Anterior Thalamic Infarction Presenting Only with Cognitive Dysfunction

A 60 year-old right-handed man developed a sudden inability to find his way home from work. Neurologic examina-tions revealed no focal neurological deficit. Neuropsychological tests showed a memory deficit especially in the visu-ospatial domain along with visuospatial and frontal-executive dysfunction. A brain MRI revealed a small but discrete lesion in the right thalamus. The majority of the lesion was confined to the dorsomedial nucleus. These results suggest that a small unilateral thalamic infarction affecting the anterior area may be sufficient to produce "strategic-infarct dementia" without focal neurological deficit.

Language Lateralization by Functional MRI: A Comparison with Wada Test-preliminary Results

PURPOSE: To evaluate the usefulness of functional MR imaging (fMRI) for the determination of languagedomi-nance and to assess differences in language lateralization according to activation task or activated area. MATERIALS AND METHODS: Functional maps of the language area were obtained during word generation tasks (noun andverb) and a reading task in ten patients (9 right handed, 1 left handed) who had undergone the Wada test. MRexaminations were performed using a 1.5T scanner and the EPI BOLD technique. The SPM program was employed for thepostprocessing of images and the threshold for significance was set at p<0.001 or p<0.01. A lateralization indexwas calculated from the number of activated pixels in three hemispheric re-gions (whole hemisphere, frontal lobe,and temporoparietal lobe), and the results were compared with those of Wada tests. The results for lateralizationof language area were compared among stimulation tasks and regions and used for calculation of lateralizationindices. RESULTS: During the Wada test, nine patients were left dominant and one patient was right dominant forlan-guage. Language dominance based on activated signals in each hemisphere was consistent with the results of theWada test in 87.5% (verb and noun generation tasks) and 90% (reading task) of patients. Language domi-nancedetermined by activated signals in the frontal lobe was consistent in 87.5%, 75%, and 80% of patients in eachstimulation task (verb generation, noun generation, and reading), respectively. The consistency rate of ac-tivatedsignals in the temporoparietal lobe was 87.5%, 87.5% and 80% of patients in each task. The mean val-ue of thelateralization index, calculated on the basis of activated signals in the temporoparietal lobe was higher thanthat in the hemisphere or frontal lobe. The verb generation task showed a higher lateralization index than thenoun generation or reading task. CONCLUSION: The lateralization index was higher in the verb generation task andin the region of the tem-poroparietal lobe than in other stimulation tasks or regions. fMRI is a potentiallyuseful non-invasive method for the determination of language dominance.

Cut-off Value of Wada Memory Score in Verbal and Visual Memory Domain

BACKGROUND: The cut-off value which determine pass or fail in Wada memory test has not been thoroughly studied, especially in regard to modality-specific memory domains. To investigate the optimal cut-off value, we tried to know what value is appropriate to forecast good postoperative memory outcome in verbal and visual memory domains. METHODS: Eighteen temporal lobe epilepsy patients underwent preoperative Wada test and pre- and postoperative neuropsychological evaluation. Corrected Wada memory score (cWMS) was calculated by ipsilateral Wada retention score divided by pre-Wada score. As a measure of postoperative memory outcome, neuropsycholgical asymmetry index (NPAI) was defined as [(postoperative memory score ? Preoperative memory score) / their mean] and analyzed in verbal and visual memory domains respectively. We made six arbitrary cut-off cWMS values (40, 50, 60, 70, 80, and 90%) and dichotomized the patients into pass or fail group at each arbitrary cut-off value. Verbal and visual NPAIs were compared statistically between two groups. RESULTS: Verbal NPAIs were significantly different between pass and fail groups at 80% cut-off value (p=0.044). Visual NPAIs were significantly different between two groups at 50% (p=0.043) and 60% (p=0.003) cut-off values. CONCLUSIONS: Higher cut-off values may be better for differentiating good and poor postoperative memory outcomes in verbal memory domain, and, in contrast, lower cut-off values may be better in visual memory domain. The verbal and visual memory in Wada test should be analyzed with different cut-off values.